search

tencent-ailab / V-Express

V-Express aims to generate a talking head video under the control of a reference image, an audio, and a sequence of V-Kps images.
2.26k stars 281 forks source link

If you have problems with slow start... #38

Closed KMiNT21 closed 5 months ago

KMiNT21 commented 5 months ago

If you encounter a very slow startup, you can safely remove the file model_ckpts\insightface_models\models\buffalo_l\1k3d68.onnx (137M)

While loading on my PC, this model significantly slows down the process (model = onnx.load(self.model_file) inside insightface/model_zoo/landmark.py).

However, this project doesn't utilize this model; it relies solely on the RetinaFace 'detection' model (det_10g.onnx). Therefore, removing the unused model eliminates the unnecessary wait during initialization.

tiankuan93 commented 5 months ago

This file is not being used and I will remove this file from the huggingface repository.